[SOLVED] perl: regex exact match using if statement

amateurscripter · 12-29-2021, 03:42 PM

Hello,

I tried few different ways to return a unique result but it's not working.

I have a script that goes through the below sample data set and then return matches based on third column. So it extracts column three and then goes through the data again to dissect each line. Notice there are dupes for each app(ie: sweep running on zpd58 and tdp58).

zpd58.vm2.config|sweep|SWEPCRS|299.352.773.112
zpd71.vm14.config|mkt|SPREADSOOT|299.352.773.113
zpd71.vm15.config|mkt|SPREADSOOT2|299.352.773.111
zpd71.vm16.config|rpt2|SOOTOT2|299.352.773.112
zpd71.vm16.config|spt2|SPAAN2|299.352.773.112
zpd72.vm14.config|rpt3|SOOTOT3|299.352.773.113
tpd44.vm2.config|sweep|SWEPCRS|288.532.235.156
tpd34.vm14.config|mkt|SPREADSOOT|288.532.235.157
tpd22.vm15.config|mkt|SPREADSOOT2|288.532.235.156
tpd28.vm16.config|rpt2|SOOTOT2|288.532.235.156
tpd41.vm16.config|spt2|SPAAN2|288.532.235.156
tpd62.vm14.config|rpt3|SOOTOT3|288.532.235.157

So when I run this part of the code, I think it's greedy and matches both SPREADSOOT and SPREADSOOT2 when it tries to match for SPREADSOOT. And notice I want to only process lines starting with "z". I tried all different variation and no luck:

if(($data =~ m/^zpd/) and ($routes =~ m/$routes\b0$data/)){

if(($data =~ m/^zpd/) and ($routes =~ m/$routes\|$data/)){

if(($data =~ m/^zpd/) and ($routes =~ m/$routes?.*$data/)){

OUTPUT:
....
The route is SWEPCRS app is sweep host: zpd58 and vm: vm2 quote:299.352.773.112
The route is SPAAN2 app is spt2 host: zpd71 and vm: vm16 quote: 299.352.773.112
The route is SPREADSOOT app is mkt host: zpd71 and vm: vm14 quote: 299.352.773.113
The route is SPREADSOOT app is mkt host: zpd71 and vm: vm14 quote: 299.352.773.113
The route is SPREADSOOT2 app is mkt host: zpd71 and vm: vm15 quote: 299.352.773.111

So when it's processing SPREADSOOT, how do I get it to not process or match against data for SPREADSOOT2?

sundialsvcs · 12-29-2021, 06:14 PM

Simplify the problem: begin with split. Since this is “pipe-delimited data,” first split it by pipe, thus creating an array, then concentrate on the one column of interest. This should completely eliminate concerns of “greediness.”

shruggy · 12-30-2021, 05:28 AM

Code:

perl -F'[|]' -le '
@a=split(/[.]/);
print"The route is $F[2] app is $F[1] host: $a[0] and vm: $a[1] quote: $F[3]" if /^z/' \
  input_data

awk is a bit more concise here:

Code:

awk -F\| '
/^z/{
  split($1,a,".")
  print"The route is",$3,"app is",$2,"host:",a[1],"and vm:",a[2],"quote:",$4
}' input_data

A shell script would do as well:

Code:

#!/bin/sh
file='input_data'

while IFS='.|' read -r h vm _ a r q
do
  case $h in z*)
    echo "The route is $r app is $a host: $h and vm: $vm quote: $q"
  esac
done <"$file"

amateurscripter · 12-30-2021, 04:36 PM

Thx guys. I'm good.